Related Experiment Videos
Direct evaluation of thermal fluctuations in proteins using a single-parameter harmonic potential
I Bahar1, A R Atilgan, B Erman
1Polymer Research Center, Bogazici University, Bebek, Istanbul, Turkey. bahar@prc.bme.boun.edu.tr
Folding & Design
|January 1, 1997
Summary
This study introduces an elastic network model to predict atomic fluctuations in proteins. The model accurately predicts temperature factors and correlations, offering a faster computational approach.
Area of Science:
- Computational Biology
- Structural Bioinformatics
- Protein Dynamics
Background:
- Investigates interactions between alpha-carbon pairs in folded proteins within 7.0 Angstroms.
- Utilizes a single-parameter harmonic potential for residue fluctuations around mean positions.
- Employs the Kirchhoff adjacency matrix to define residue proximity in protein structures.
Purpose of the Study:
- To develop and validate an elastic network model for predicting atomic fluctuations and correlations in proteins.
- To assess the model's accuracy and computational efficiency compared to existing methods.
Main Methods:
- Developed an elastic network model based on the Kirchhoff adjacency matrix.
- Calculated auto-correlations and cross-correlations of atomic fluctuations from the inverse Kirchhoff matrix.
- Applied the model to 12 X-ray protein structures of varying sizes.
Main Results:
- Accurately predicted temperature factors for C alpha atoms across diverse protein structures.
- Efficiently characterized cross-correlations between atomic fluctuations.
- Results closely matched those from normal mode analysis coupled with energy minimization.
Conclusions:
- The proposed elastic network model satisfactorily describes correlations in atomic fluctuations.
- The method offers a significant computational speed advantage, being at least one order of magnitude faster than conventional molecular approaches.
- This simplified model provides an efficient tool for analyzing protein dynamics.